Tables
Table 1 The average performance of the models across 10
crossvalidated folds of the training dataset calculated for random and
background weighting (BkWt) schemes. The threshold-dependent measures
were calculated given a 10 percentile of suitability score at the
training occurrence points.